20 BULLETIN 218, U. S. DEPARTMENT OF AGRICULTURE. 
quite so high, while those following sweet clover have been still 
lower. 
The average yield on alfalfa sod has been the lowest in the series. 
On both brome-grass and clover sods the average yields have been 
practically the same as after small grains. 
The yield of oats following wheat on both spring-plowed and fall- 
plowed land appears to be better than when following oats on land 
so prepared. 
Little difference is to be observed in the average results following 
either spring or fall plowing of small-grain stubble for oats. There 
are differences that develop with differences in seasons, but on the 
whole it would appear from the evidence at hand that the time for 
plowing for oats at this station would be determined chiefly, if not 
solely, by the matter of economy and convenience in doing the 
work. 
When the cost of preparation by the different methods under 
study is taken into consideration, it is seen that this cost, rather 
than differences in yield, is the determining factor. The greatest 
average profit, $5.55 per acre, has been realized from oats on disked 
corn ground. Oats on spring-plowed and fall-plowed land have 
been productive of nearly equal profits of about $2 less. The 
slightly increased yield by summer tillage has not been sufficient to 
meet the increased cost of the method and profits from it have 
fallen to 24 cents per acre. Green manuring has been responsible 
for an average loss of $3.83 per acre. 
HETTINGER FIELD STATION. 
The soil at the field station at Hettinger, N. Dak., is a heavy clay 
loam. The seasons during which the work has been carried on have 
been such that the results of soil-moisture study are not yet conclu- 
sive in determining the proportion of water that can be stored in the 
soil and recovered by a crop. It is probable, however, that the depth 
of feeding is not limited by any physical peculiarity of the soil and 
that the supply of available water that can be stored is large. It is 
reasonable, therefore, to expect that on this soil the maximum effect 
will be realized from methods of tillage calculated to store water. 
The results of three years of fair production are available from the 
Hettinger station. Records for this length of time are not so valu- 
able an index to methods of production as the longer records at other 
stations in the State. It appears evident, both from the records and 
from field observations, that they are complicated somewhat by soil 
differences. While some of these differences are recognized in their 
manifestations, their nature has not been satisfactorily determined. 
In the study as here arranged, the most difficulty is offered by the 
two unduplicated plats of oats following oats, one prepared by spring 
